• S4062-PT (Fig. 12.99)

The S4062 was intended to be an improvement over the S4061. The major modification is a longer run of laminar flow designed into the S4062. But as the data shows, this design change has not lead to improved performance. Rather, the S4062 has higher drag owing to a larger laminar separation bubble. It is interesting to note that this problem is also predicted by the ISES code which was not available when the S4062 was designed.

Also see: S4061, SD6080, DAE51 Digitizer plot: Fig. 10.37 Polar plot: Fig. 12.99

Thickness: 9.53% Camber: 4.14%


Although originally intended as a cross-country airfoil, the S4061 has since demonstrated it’s versatility under a variety of conditions. At a recent AM A Nationals (the NATS), Paul Carlson, founder of Off the Ground Models, flew his newly designed and kitted PRODIGY sailplane to a first place finish in the 2-Meter class, a second in Standard, and a third in Unlimited. To date over 3500 PRODIGY kits have been produced, and the popularity of the S4061 has grown proportionally.

• S4061A-PT (Fig. 12.93)

• S4061B-PT (Fig. 12.94)

The version A model of the S4061 is inaccurate. The more accurate version В shows lift and drag characteristics much different than those of the A. As was amply demonstrated in other examples, these differences in the data come as no surprise.

• S4061B-PT u. s.t. xjc = 45%,hjc = .17%,v>/c = 1.0% (Fig. 12.95)

• S4061B-PT u. s.t. xjc = 45%, Rn = 150,000 (Fig. 12.96}

• S4061B-PT u. s.t. xjc = 45%, Rn = 150,000 and 300,000 (Fig. 12.97)

As shown in Fig. 12.95, a trip with height of 0.17% reduces drag for Rn less than 150k. Figures 12.96 and 12.97 show that a two-dimensional trip produces the same result as zig-zag tape (type A).

Also see: SD6080, DAE51, S4062, SD7037, E387, E193, E214 Digitizer plot: Figs. 10.35, 10.36 Polar plot: Figs. 12.93-12.97 Lift plot: Fig. 12.98

Thickness: 9.60% Camber: 3.90%


• S3021A-PT (Fig. 12.90)

• S3021B-PT (Fig. 12.92)

The S3021 has a bubble ramp longer than the transition ramp on the E205, and the resulting reduction in drag is dramatic. Certainly some of the reduction is because the S3021 is thinner than the E205, but, from the shape of the airplane polar shown in Fig. 5.9, it is clear that most of the drag reduction comes from better management of the laminar separation bubble.

Two models of the S3021 were built; version A is the more accurate one. Version В has a reflexed trailing edge which produces a steeper pressure recovery on the upper surface—a change which leads to a higher bubble drag.

Fig. 4.4 compares the data from the S3021A-PT and the E2Q5B-PT (Figs. 12.90 and 12.14 respectively) and shows that the drag of the S3021A-PT is almost everywhere lower than that of the E205B-PT. The ISES code predicted a similar drag reduction when it was used to compare the nominal airfoils.

Also see: E205, S3014, S3010, SD7080, SD7084

Digitizer plots: Fig. 10.33, 10.34

Airfoil comparision plot: Fig. 11.9

Polar plot: Figs. 12.90, 12.92

Lift plot: Fig, 12.91

Aircraft polar: Fig. 5.9

Thickness: 9.47% Camber: 2.96%

S3014 and S3016

• S3014-PT (Fig. 12.87)

• S3016-PT (Fig. 12.89)

The S3010, S3014 and S3016 are quite similar aero dynamically since the latter two were derived from the first with only minor modifications. Although the S3016 has a slight edge over the S3010 and S3014 at high speed and the S3014 looks best at a Rn of 60k, none of the three has a decided advantage over the others.

Also see: S3021, E205, SD7080 Digitizer plot: Figs. 10.31, 10.32 Polar plot: Figs. 12.87, 12.89 Lift plot: Fig. 12.88

S3014 Thickness: 9.46% Camber: 2.57%

S3016 Thickness: 9.52% Camber: 2.09%


• S3010-PT (Fig. 12.85)

The S3010 is the first example of a Selig airfoil that has a lift range much like the E193 and E205 airfoils. Other examples include the flat-bottom S3014, S3016, and S3021. All these designs took advantage of the increased understand­ing of the laminar separation bubble, and as a result the high-drag bulges in the middle of the low-i? n polars are virtually gone. Instead, the polar and its edges are more rounded, and in most planes the drag of the S3010 is lower than the E205. It is interesting to note that in many respects the S3010 performance characteristics are very similar to the DF101, although as shown in Fig. 11.8, a comparison of the profiles actually tested reveals that the S3010 and DF101 shapes are very different.

Also see: S3021, CLARK-Y, DF101, S3014, S3016, E205

Digitizer plot: Fig. 10.30

Airfoil comparision plot: Fig. 11.8

Polar plot: Fig. 12.85

Lift plot: Fig. 12.86

Thickness: 10.32% Camber: 2.82%


As mentioned in the discussion of the AQUILA airfoil, its high-speed per­formance is severely compromised by its high-camber. The S2091 was designed primarily to be an improvement over the AQUILA by extending the polar to lower lift, while maintaining the AQUILA’s low-speed, high-lift characteristics. Details of how this was achieved through airfoil design may be found in Refer­ence 23.

• S2091A-PT (Fig. 12.79)

• S2091B-PT (Fig. 12.80)

Data on two models are shown. The first model, version A, was found to be too thin as verified by hand-held templates. In addition, the leading edge was rough in certain areas from the fiberglass beneath the paint. Undoubtedly this influenced transition in a manner similar to that of a trip strip. The model was later re-contoured and given the В designation. Only the В version was digitized for coordinates.

The effect of the roughness of version A as compared with the smooth version В is to decrease the drag at 60k. Smaller improvements are found for 100k, while no improvement exists at 200k.

Referring to the В version data (Fig. 12.80), the goal of extending the low – lift end of the polar beyond that of the AQUILA airfoil has been achieved. Futhermore, it comes as some surprise that at Rn’s above 60k the maximum lift coefficient is increased by 0.1 over that of the AQUILA. In summary, the S2091 is an advance in performance over the AQUILA, but this comes mostly through the relaxation of the flat-bottom requirement. Of course a flat lower surface does not necessarily mean the airfoil is deficient, as illustrated by the performance of the DF101 and the nearly flat-bottom S3021.

• S2091B-PT Gurney Flap type A (Fig. 12.81)

• S2091B-PT Gurney Flap type В (Fig. 12.82)

• S2091B-PT Gurney Flap type C (Fig. 12.83)

The S2091B-PT was tested with a so-called Gurney flap, often used on facing car wings which develop a download to increase traction. As can be seen in Fig. 5.5, a Gurney flap is a simple, thin tab on the order of 1% chord, which is perpendicular to the lower side of the airfoil at the trailing edge. For these tests the tab was 0.017% chord thick (0.002 in) brass shim stock with a length of 0.6% chord for type A, 1.2% for B, and 2.6% for C. •

For the type A Gurney flap, the whole polar is shifted upwards in lift coeffi­cient by 0.1—an impressive result for such a simple modification. Similar results are found for type B, but diminishing returns begin to appear for type C. It can be reasonably expected that the efficiency of the small 0.6% and 1.2% Gurney flaps in increasing lift is not unique to the S2091. (Note that there is a small drag penalty associated with the Gurney flap.)

Also see: AQUILA, E214, SD7037, SD7032 Digitizer plot: Fig. 10.29 Polar plot: Figs. 12.79-12.83 Lift plot: Fig. 12.84

Thickness: 10.10% Camber: 3.91%


• S2055-PT (Fig. 12.78)

The S2055 is a slightly thinned and decambered S2048. The results of this small perturbation in shape are marginally lower drag and a downward shift of the lift range. As with other F3B airfoils, flaps are necessary to fully realize the capabilities of the airfoil.

Of all the airfoils tested, the S2055-PT had the lowest drag at 300k. Although several attempts were made to reduce the drag further through the use of trips, none worked. Instead, the drag was increased except when the trip was far aft, and then it had no effect.

Although this is the lowest drag airfoil tested and should do very well on sailplanes where speed is paramount, the 8% thickness offers a significant con­struction challenge.

Also see: SD2030, S2048, RG15, HQ2/9, SD8000 Digitizer plot: Fig. 10.28 Polar plot: Fig. 12.78

Thickness: 7.99% Camber: 1.66%


• S2048-PT (Fig. 12.74)

• S2048-PT with trips (Fig. 12.75)

• S2048-PT misc. trips (Fig. 12.76)

The S2048 was originally presented at the 1985 MARCS Symposium held in Madison, Wisconsin and has been used on the Synergy F3B sailplanes that have represented the U. S. at the World F3B Championships held in recent years (1987 and 1989). The airfoil is a “redesigned” HQ2/9 with slightly longer bubble ramps on the upper and lower surfaces (i. e. more gradual pressure gradients). Because the performance improvement due to changing these ramps was expected to be small in any case, more accurate models than the ones we tested would be re­quired to detect and measure it. And even if it could be measured, actually realizing or maintaining that difference would be difficult. Nonetheless the pos­sibility of bettering existing F3B airfoil designs may still yield to investigation at another time.

The effects of trips on the upper and lower surfaces are shown in Fig. 12.75; results of miscellaneous trip locations are shown in Fig. 12.76. Although no attempt was made to systematically study the effects of trip location on this airfoil, it does seem that a slight improvement in high-speed performance can be obtained through the use of an upper-surface trip around 60-70% chord combined with a lower-surface trip near 60%. As with the RG15 and HQ2/9, trips nearer the leading edge can be expected to produce lower drag at lower i£n’s.

Also see: HQ2/9, RG15, S2055, SD8000, SD2030

Digitizer plot: Fig. 10.27

Airfoil comparision plot: Figs. 11.4, 11.7

Polar plot: Figs. 12.74-12.76

Lift plot: Fig. 12.77

Thickness: 8.63% Camber: 1.94%


The original group of Selig airfoils23 was designed with the aid of both the Eppler and Somers Computer Code for the Design and Analysis of Low-Speed Airfoils20 and the experimental results of Althaus9. The approach taken was to compare the theoretical and experimental results of several airfoils, then to determine from these comparisons what factors (in the velocity distribution and boundary layer development) were needed to produce good, low-iZn airfoils. These factors were subsequently incorporated into the design of several new airfoils for RC sailplanes.

A common feature of most of the Selig airfoils is a long, gradual pressure recovery region on the upper surface called a bubble ramp. As a result of this ramp, the upper-surface transition point moves forward slowly and continuously with increasing angle of attack. This feature usually gives the polars an appear­ance more closely resembling those measured at high Rn. These are general design principles. Surprisingly, even today the details of how “gradual” and how “slowly”, and how best to achieve this are topics of research.

The performance of the Selig airfoils is often markedly different from one to the next. This is because the airfoils were designed to have a variety of char­acteristics in the hope that trends would emerge, leading ultimately to a better understanding of low-Rn airfoils. Implicit in this approach is the recognition that low-Rn airfoil design is still in the early stages of development.

The S-designation is also used by Somers; however, Selig uses four digits in the airfoil name, while Somers uses only three.


• RG15-PT (Fig. 12.68)

• RG15-PT u. s.t. xjc = 20%, h/c – .17%, w/с = 1.0% (Fig. 12.69)

• RG15-PT u. s.t. xjc = 40%, h/c = ,17%,u;/c = 1.0% (Fig. 12.70)

• RG15-PT u. s.t. xjc = 60%, h/c = .17%, w/c = 1.0% (Fig. 12.71)

• RG15-PT u. s.t. xjc = 70%, h/c = .17%, w/c — 1.0% (Fig. 12.72)

The RG15 was tested extensively with trips because the performance of the untripped model (shown in Fig. 12.68) was relatively good. We asked ourselves: when one starts with a “good” airfoil (the RG15-PT), can trips still make modest improvements in performance? Note that the question is specific to this airfoil and cannot be generalized.

It is instructive to begin with the trip at 70% chord shown in Fig. 12.72. At this location, it has virtually no effect since it is downstream of laminar separation. In other words, it is either inside the laminar separation bubble or immersed in the turbulent boundary layer.

At 60% chord, the trip causes the 150k and 200k curves to spread apart. At 40% chord, the trip efficiently trips the boundary layer for Rn of 150k at the lower lift coefficients. As observed before, the higher-i£n polars overlap since the boundary layer is tripped too soon, producing more drag than that found for the lower Rn’s (150k in this case).

At 20% chord, there is even more overlap at 300k. But the drag at 100k has decreased significantly from the untripped case as the trip moves forward. As is true with the HQ2/9, a trip should be employed for Rn less than 150k-200k. The location will depend on the average local chord Rn.

Also see: HQ2/9, S2048, SD8000, SD2030 Digitizer plot: Fig. 10.26 Airfoil comparision plot: Figs. 11.3, 11.6 Polar plot: Figs. 12.68-12.72 Lift plot: Fig. 12.73

Thickness: 8.92% Camber: 1.76%